Novita AI
A finite allowance for new accounts; it does not recur.
https://api.novita.ai/v3/openaiModels mentioned
inclusionai/ling-3.0-flash-finbunnyqwen/qwen3.5-plusqwen/qwen3.6-plusinclusionai/ling-3.0-flash-vlinclusionai/ling-3.0-flash-santedev/glm46Equivalent paid value
How this was valued: The published allowance or a defensible paid comparison is not precise enough to calculate.
The decisive terms remain authenticated-only or unbounded: the new-user voucher's amount and expiry are disclosed only in the signed-in dashboard (third-party $0.50 reports are not first-party evidence), and while the live catalog now lists genuinely zero-priced LLM routes with documented per-request output ceilings (inclusionai/ling-3.0-tiny at 32,768 and nex-agi/nex-n2-pro at 262,144 max output tokens; the latter priced $0.25/$1.00 per 1M on OpenRouter), Novita's public rate-limit documentation covers only image and video models and publishes no LLM requests-per-minute or tokens-per-minute ceiling, so no finite period envelope can be built for the zero-priced routes without inventing a request rate.
Limits and terms
- Signup Voucher Amount
- dashboard only
- Recurrence
- No
- Caveat
- The finite new-user voucher is followed by prepaid top-ups; promotional vouchers can expire.
What happens to your prompts?
Default inference content is transient and excluded from training and service improvement, subject to narrow legal and support exceptions.
- Plan Scope
- Novita AI marketplace inference APIs and web services; storage products, sandboxes, support, and legal exceptions have separate persistence.
- Prompt Retention
- Default inference Content is retained only for request processing and delivery. Exceptions permit retention where law requires or where necessary for service delivery or technical support.
- Response Retention
- Same default ZDR and support/legal exceptions as prompts.
- Ordinary Logging
- Account data persists for the account lifetime plus seven years; technical logs can persist two years, transaction records seven years, and communications three years without turning inference content into ordinary logs.
- Model Training
- Novita states it does not use Content to train its own models by default.
- Product Improvement
- The terms state Content is not used to improve services by default; de-identified personal and technical information may still be used indefinitely for research and statistics.
- Human Or Operator Access
- Content is not logged for human review under default ZDR, except where needed for service delivery, technical support, law, or security. Automated safety screening is always permitted.
- Subprocessors And Routing
- Novita operates a model marketplace and uses service providers; underlying model licenses and optional storage or sandbox products can add separate rules.
- Deletion Controls
- Ordinary inference content is discarded after delivery. Account and stored-product data can be deleted subject to the lengthy legal, tax, support, and technical-log schedules.
- Caveat
- The ZDR promise applies to inference Content, not persistent storage products, paused sandboxes, support submissions, account records, or technical metadata.
Governing documents
- Terms Of Service
- https://novita.ai/legal/terms-of-service
- Privacy Policy
- https://novita.ai/legal/privacy-policy
- Acceptable Use Policy
- https://novita.ai/legal/acceptable-use-policy
Primary sources
Before you build with Novita AI
Read the classification narrowly
This record has current first-party evidence for some zero-cost hosted inference. The label describes the bounded offer supported by evidence on 2026-09-19; it does not guarantee permanence, production suitability, uptime, latency, model quality, or access from every region.
Resolve the live model route
This snapshot records 7 model IDs. Match the exact ID against the provider's live catalog before using it in code; zero-price routes and aliases can rotate while an older documentation page remains online. The recorded base endpoint is https://api.novita.ai/v3/openai, but the provider's current API reference remains authoritative for paths, authentication, and request shape.
Confirm account and billing boundaries
The public evidence does not settle whether an account is required. Confirm the signed-in flow before treating anonymous or credential-free access as available. The payment-card requirement is conditional or not clearly documented. Treat signup friction and billing exposure as unresolved until the account flow confirms them. The equivalent paid value shown here prices a documented allowance where possible; it is not cash, guaranteed savings, or protection from overage.
Re-read the prompt policy for your plan
This record classifies the reviewed handling as “Private”: Default inference content is transient and excluded from training and service improvement, subject to narrow legal and support exceptions. Provider policies can distinguish free, paid, enterprise, opted-in, and feature-specific traffic, so verify the governing terms for the exact account and route that will receive your data.
Follow the evidence, then re-check it
The record links 9 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest governing document or live catalog when sources disagree, and submit a correction when a provider changes a material term.
Match the quota to the workload shape
Translate the published allowance into the traffic pattern you actually expect instead of comparing headline totals alone. A daily token pool can look generous while a low requests-per-minute or concurrency ceiling blocks interactive bursts; a high request limit can still fail a long-context job when input, output, or per-request tokens are capped. Separate prompt tokens from generated tokens, include retries and tool calls, and test the largest realistic payload. If the provider documents more than one limit window, the tightest window at your peak load is the practical ceiling. Research, experimental, and community access can also carry eligibility or fair-use constraints that cannot be modeled as a simple number.
Plan fallback without changing the rules
A fallback should preserve more than API syntax. Confirm that the substitute route supports the required modality, context length, streaming behavior, structured output, tools, and safety controls, then compare its prompt-retention and training terms. An OpenAI-compatible request shape does not make providers operationally or contractually equivalent. Decide which errors may trigger a retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. If deterministic output matters, record the model revision and sampling settings; rotating aliases and free-model pools can change behavior even when the endpoint remains available.
Monitor the offer as a dependency
Capture the model ID, response model field, rate-limit headers, usage fields, latency, HTTP status, and any provider request identifier for each test. Watch for authorization failures, quota exhaustion, catalog removal, policy revisions, and dashboard balance changes as separate failure modes. Re-check the provider’s live catalog and governing pages on a schedule proportionate to the workload’s importance, and keep the dated sources that supported your decision. Free capacity is especially suitable for prototypes, evaluations, fallbacks, and bounded workloads when the application can tolerate change; a production dependency still needs observability, an exit path, and an owner responsible for re-verification.
Test one complete request before scaling
Start with the smallest permitted request using the exact credential, model ID, endpoint, and account type you intend to deploy. Record the HTTP status, response headers, usage fields, latency, and any dashboard balance change. Then exercise the failure path: an invalid model, an exhausted quota, or a rate-limit response should fail clearly without silently switching to a paid route. If the provider supports streaming or tool calls, validate those features separately because a free model can expose a narrower capability set than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and avoid sending sensitive data until the observed route matches the reviewed data agreement.